cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Irpan Adiputra pardosi
Contact Email
irpan@mikroskil.ac.id
Phone
+6282251583783
Journal Mail Official
sinkron@polgan.ac.id
Editorial Address
Jl. Veteran No. 194 Pasar VI Manunggal,
Location
Kota medan,
Sumatera utara
INDONESIA
Sinkron : Jurnal dan Penelitian Teknik Informatika
Published by Politeknik Ganesha Medan
ISSN : 2541044X     EISSN : 25412019     DOI : 10.33395/sinkron.v8i3.12656
Core Subject : Science,
Computer Science & IT
Scope of SinkrOns Scientific Discussion 1. Machine Learning 2. Cryptography 3. Steganography 4. Digital Image Processing 5. Networking 6. Security 7. Algorithm and Programming 8. Computer Vision 9. Troubleshooting 10. Internet and E-Commerce 11. Artificial Intelligence 12. Data Mining 13. Artificial Neural Network 14. Fuzzy Logic 15. Robotic
Arjuna Subject : Ilmu Komputer (semua kategori) - Visi Komputer dan Pengenalan Pola
Articles 1,361 Documents
 133   134   135   136   137 
K-Means and Fully Connected Neural Network for Child Nutritional Status Classification Rismayanti; Sumi Khairani
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16400

Abstract

Stunting remains a persistent child nutrition problem because delayed growth is closely related to long-term health, cognitive, and productivity risks. Manual interpretation of anthropometric measurements using the World Health Organization Z-score standard is clinically valid, yet it becomes inefficient and error-prone when routine records are processed in large numbers. This study develops a child nutritional status classification model by combining K-Means clustering and a fully connected neural network for early identification of stunting, underweight, and wasting. The dataset consisted of toddler anthropometric records from 2021-2024 with sex, age, body weight, and body height attributes. The data were cleaned, standardized, transformed into Z-score indicators, and grouped into 27 clusters representing possible combinations of nutritional status profiles. Cluster membership was then used with Zlen, Zwei, and Zwfl features in a multi-head fully connected neural network. Evaluation on 82 held-out samples showed accuracy values of 91.46% for stunting, 93.90% for underweight, and 98.78% for wasting. Weighted precision, recall, and F1-score were consistently high across the three outputs, while the training curves indicated stable learning without strong overfitting. The proposed hybrid model improves the reliability of child nutrition classification and can support a web-based decision support system for data-driven nutritional screening and intervention planning.
Target-Characteristic-Aware Forecasting of Fuel–Equity Rolling Correlations: Evidence That PCA and Persistence Outperform Hybrid Deep Learning Models Chrisnawan Prastya Atmaja; Ferian Fauzi Abdulloh
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16410

Abstract

Background: Forecasting dynamic fuel–equity market correlations is important for financial forecasting because fuel price movements can affect market risk, investor sentiment, and cross-market stability. However, most previous studies focus on direct price or volatility prediction, while the forecasting of rolling correlations between multiple fuel types and global equity indices remains less explored. Objective: This study analyzes and predicts time-varying correlations between global fuel prices and major stock market indices by comparing baseline, PCA-based, standalone deep learning, and hybrid deep learning models. Methods: The proposed framework applies data preprocessing, return transformation, 60-day rolling correlation construction across three fuel variables (petrol, diesel, LPG) and five stock indices (S&P 500, NASDAQ, FTSE 100, Nikkei 225, IHSG/JKSE), normalization, Principal Component Analysis (PCA), sequence generation, and comparative evaluation of 15 forecasting models. Model performance was measured using RMSE, MAE, R-squared (R²), and Directional Accuracy. Results: The PCA Model achieved the lowest RMSE of 0.016909 and the highest R² of 0.967045, while the Persistence Model produced the lowest MAE of 0.002293 and the highest Directional Accuracy of 97.767280%. Hybrid deep learning models showed higher errors and negative R² values, indicating weaker performance on smooth and persistent rolling correlation targets. Conclusion: The findings show that architectural complexity does not necessarily improve forecasting performance when the target series is smooth and persistent. This study contributes empirical evidence and a replicable comparative framework showing that model selection in financial time-series forecasting should consider target characteristics, particularly smoothness and temporal persistence, rather than relying solely on complex deep learning architectures.
Prioritizing Governance of the COBIT 2019 DSS02 Domain Based on Incident Management Data Lahan Adi Purwanto; Achmad Fauzan
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16421

Abstract

Information technology (IT) governance plays a crucial role in ensuring service quality, and COBIT 2019 provides a structured framework for incident management through the DSS02 domain. However, DSS02 implementation is commonly evaluated using surveys or capability assessments that are subjective, difficult to reproduce, and challenging to verify independently, despite the availability of detailed operational event logs in IT service management platforms. This study proposes an event-log-based approach to identify governance improvement priorities within the COBIT 2019 DSS02 domain. Five operational indicators were developed, namely the Process Complexity Index (PCI), Operational Interaction Index (OII), Reassignment Indicator (RI), Reopened Incident Indicator (RII), and Delay Severity Index (DSI). These indicators were mapped to DSS02 subprocesses and integrated into a Governance Priority Score (GPS), which was evaluated using local sensitivity analysis and Spearman correlation analysis on a ServiceNow event log containing 141,707 events and 24,918 incidents. The results identified DSS02.04 (Investigate, Diagnose and Resolve) as the highest governance priority (GPS = 5.06), followed by DSS02.02 (4.27) and DSS02.03 (1.14). Sensitivity analysis showed that moderate (±10%) changes in indicator weights did not alter the priority ranking, while correlation analysis revealed a strong relationship between reassignment frequency and incident modification activity (ρ = 0.67). The proposed approach provides an objective, reproducible, and auditable data-driven prioritization mechanism that complements conventional survey-based governance assessment using operational event logs.
Adaptive Hybrid Model for Academic Performance Prediction and Learning Strategy Recommendation Nenna Irsa Syahputri; Hasdiana Hasdiana
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16430

Abstract

Academic performance prediction is important for helping lecturers identify student learning needs before academic problems become difficult to address. However, students differ in learning preferences, engagement, prior knowledge, and academic achievement, making uniform learning strategies less effective. This study proposes an adaptive hybrid model for academic performance prediction and learning strategy recommendation by integrating K-Means Clustering, Simple Additive Weighting, and an Artificial Neural Network. The dataset consists of 74 student samples containing VARK learning preferences, engagement scores, pretest scores, and GPA-like academic indicators. After data cleaning, median imputation, and standard scaling, K-Means was applied to segment students into five learning profiles. Cluster centroids were then transformed into three decision criteria, namely Engage, Retention, and Effort. Simple Additive Weighting was used to rank three learning strategies: Micro-video Learning, Quiz Drill Practice, and Peer Discussion. The resulting recommendation labels were used together with the academic features to train an Artificial Neural Network for performance prediction and strategy classification. The evaluation showed that both models achieved an unrounded accuracy of 99.63%, while the rounded classification report displayed nearly perfect precision, recall, and F1-score. These findings indicate that the proposed integration can support data-driven adaptive learning decisions. Nevertheless, the high performance should be interpreted carefully because the dataset is limited and comes from a single institutional context. Further validation with larger, more diverse datasets is required to confirm generalizability.
AI-Based Pharyngitis Detection Expert System Using the Certainty Factor Method Emy Susanti; Robby Cokro Buwono; Dison Librado; Faiz Rifki Syuhada
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16463

Abstract

Pharyngitis, an inflammation of the pharynx commonly known as a sore throat, is one of the most frequent complaints in Indonesian primary health care, yet the uneven distribution of physicians, especially in remote regions, often delays timely triage. This study develops an Android-based expert system that provides an early-screening indication of two types of pharyngitis, acute and chronic, using the Certainty Factor (CF) method. Rather than treating the disease application itself as the main contribution, the novelty lies in the validation framework: a dual-source certainty model that keeps expert-elicited rule weights separate from user-reported symptom confidence, and a class-level accuracy evaluation, including a confusion matrix, sensitivity, specificity, precision, recall, and F1-score, that is rarely reported alongside comparable Certainty Factor systems. Knowledge was acquired from a general practitioner and clinical references and encoded as thirteen symptoms, two disease classes, and a rule base of IF-THEN rules with expert certainty weights. User certainty is captured through six linguistic terms and combined with the expert CF values using the single-evidence formula CF[H,E] = CFuser x CFexpert and the parallel combination formula. The application was built with Android Studio and a local SQLite knowledge base so that it operates entirely offline. A worked example involving five symptoms of acute pharyngitis produced a combined certainty of 0.9890, or 98.90%, illustrating a high-confidence screening output. The system's screening conclusions were compared with a general practitioner's diagnosis on 30 patient cases, agreeing in 28 cases for an overall agreement accuracy of 93.33% (class-level sensitivity of 94.12% for acute and 92.31% for chronic pharyngitis); because this figure reflects agreement with a single practitioner rather than a laboratory-confirmed reference standard, it is reported as an agreement accuracy rather than a universal clinical accuracy. Black-box testing confirmed that all functional features operated as intended. The results indicate that the Certainty Factor method can quantify screening uncertainty effectively and that the resulting offline mobile application can serve as an accessible early-screening aid for the public and a decision-support tool for paramedics, complementing rather than replacing a professional medical examination.
SC-Literature Intelligence: A Retrieval-Augmented Generation Framework for Multi-Category AI Literature Synthesis in Supply Chain Setio Basuki; Amelia Khoidir; Muhammad Ilham Perdana; Muhammad Daffa Nugraha; Masatoshi Tsuchiya
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16477

Abstract

This paper develops SC-Literature Intelligence, a retrieval-augmented generation (RAG) framework for research synthesis of scientific literature on artificial intelligence (AI) in the supply chain domain. The study addresses the fragmentation of scientific findings, which makes cross-document understanding difficult by supporting four categories of literature-analysis queries: trend analysis, gap detection, comparative synthesis, and evidence-based question answering (QA). The primary novelty lies in introducing a category-aware research synthesis framework capable of evaluating RAG performance across multiple literature-analysis tasks rather than conventional question answering. The framework is built from Scopus-indexed abstracts through pre-processing, chunk-based embedding using BGE-M3 and LaBSE, vector storage, semantic retrieval, and prompt-guided generation evaluated using the RAGAS framework across 640 experimental runs. The results show that BGE-M3 consistently outperforms LaBSE on all RAGAS indicators with the best configuration (chunk size 64, Top-K 5) achieving scores between 0.722 and 0.856 across faithfulness, answer relevancy, context precision, and context recall. Gap detection emerges as the best-supported query category, whereas comparative synthesis remains the most challenging. Failure analysis further reveals that retrieval-stage issues dominate over generation-stage issues, identifying embedding quality as the primary bottleneck. These findings demonstrate that category-aware RAG-based synthesis can support structured, evidence-grounded literature analysis in the supply chain AI domain.
Predictive Analytics for Energy Consumption of Autonomous Mobile Robot Using Hybrid ARIMA-XGBoost Pipit Anggraeni; Wahyu Adhie Candra; Surya Dharma Jatnika; Noval Lilansa; Adhitya Sumardi Sunarya; Nur Jamiludin Ramadhan; Andri Wiyono
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16484

Abstract

The deployment of autonomous mobile robots in smart manufacturing and intralogistics has grown rapidly, yet current battery management systems can only monitor real-time charge levels without predicting future energy consumption. This reactive limitation risks mid-mission battery depletion and production disruption. The study presents a predictive analytics system for the Polebot autonomous mobile robot integrated with a Robot Operating System 2 data historian pipeline and an InfluxDB time-series database. The objective is to evaluate whether an autoregressive time-series model or a gradient-boosted machine learning model better suits different operational conditions, specifically constant-velocity static operation versus acceleration-heavy dynamic operation. Data were collected from three sensor sources across six operational protocols covering baseline, high-load, stop-and-go, creep, burst acceleration, and mixed conditions, yielding 10,800 synchronized data points at 1 Hz after resampling. Results show that the Autoregressive Integrated Moving Average model with parameters (2,1,3) achieves a Mean Absolute Error of 1.047% and a symmetric Mean Absolute Percentage Error of 1.74% for battery State of Charge prediction under static conditions. Extreme Gradient Boosting achieves a Mean Absolute Error of 0.022 watts for motor power prediction, 136 times more accurate than the time-series model for the same variable. The proposed Condition-Based Temporal Switching framework was validated on 1,803 data points and autonomously produced 265 model transitions during a 30-minute mixed operational test, with static conditions comprising 83.9% of the validation window. Adaptive model selection outperforms single-model strategies for energy prediction in autonomous mobile robot platforms.
Analysis of Image Augmentation Effects on MobileNetV2 for Gastroesophageal Reflux Disease Endoscopy Image Classification Dinda Chesar Putri Ramadhani; Risqy Siwi Pradini; M. Syauqi Haris
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16510

Abstract

Gastroesophageal Reflux Disease (GERD) is a common digestive tract disorder, and endoscopic image-based diagnosis still faces challenges due to variations in lighting, image capture angles, and limited data that can cause deep learning models to overfit and reduce generalization capabilities. While previous studies frequently prioritize overall accuracy, this study distinguishes itself by focusing on increasing recall as a critical indicator of clinical sensitivity for positive case detection. This study aims to analyze the effect of applying image augmentation on the performance of GERD endoscopic image classification using a lightweight MobileNetV2 architecture, offering a practical solution for resource-constrained clinical settings. The research methods include data collection, class mapping, 80:20 data split, and standardized preprocessing stages. Two training scenarios were compared: without augmentation and with strategic on-the-fly augmentation applied through a transfer learning approach. Evaluation was carried out using accuracy, precision, recall, F1-score, and ROC-AUC. The results showed that augmentation improved model performance, with accuracy increasing from 76.72% to 78.06%, recall increasing from 79.33% to 85.27%, F1-score increasing from 77.86% to 80.04%, and AUC increasing from 0.8317 to 0.8481. These findings indicate that the augmentation strategy effectively improves the generalization of the lightweight model, significantly reducing missed diagnoses to support early detection of GERD. In addition, the results demonstrate that applying augmentation techniques can help the model learn more robust visual features from limited medical image datasets. This approach also contributes to reducing overfitting and improving model reliability in practical clinical image analysis scenarios.
Incremental Effects of Augmentation Strategies on Pre-Augmented Biomedical Waste Detection Using YOLOv11n Jevri Tri Ardiansah; Ahmad Kholish Fauzan Shobiry; Anik Nur Handayani; Mohammad Muzayyin Amrulloh; Taiga Haruta
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16531

Abstract

Biomedical waste carries infectious and hazardous risk that makes accurate automated sorting valuable, and object detection offers a path toward automation, yet published studies rarely measure how much online data augmentation contributes once training images carry offline augmentation and mean Average Precision alone can conceal how augmentation shifts the balance between missed detections and false alarms.This study measures the incremental effect of Mosaic, MixUp, and Copy-Paste augmentation on YOLOv11n trained for biomedical waste detection, and verifies whether each augmentation executes as configured. We designed a 2×2×2 factorial ablation across eight configurations, trained YOLOv11n three times per configuration with different random seeds on a 14-class biomedical waste dataset and evaluated each run's best checkpoint on a held-out test set using mean Average Precision, precision, recall, and a corrected confusion matrix retaining the background class. We verified framework behavior through prediction-level identity, loss-level comparison, and instrumented tracing. Configurations combining Mosaic and MixUp raised mean Average Precision by 0.0053 over baseline, MixUp alone raised recall from 0.8950 to 0.9071, and Mosaic with MixUp raised precision to 0.9650. False negatives outnumbered false positives three to one across every configuration, identifying class-versus-background rather than inter-class confusion as the dominant failure mode. Copy-Paste showed zero measurable effect once verified, consistent with its restriction to segmentation tasks in the underlying framework. Online augmentation produces modest but distinguishable, metric-specific gains on pre-augmented biomedical waste data, and verifying framework behavior before attributing results to an augmentation strategy is necessary for reliable reporting.
Fraud Detection in E-Commerce Transactions Using Autoencoder Anomaly Scoring and XGBoost Classification Angelina Law; Stephen Sanjaya; Ferico Carvius Wivano; Osman Renjiro Giawa; Yennimar Yennimar
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16551

Abstract

The rapid expansion of e-commerce platforms has intensified the risk of digital transaction fraud, which is particularly challenging to detect due to highly imbalanced datasets where fraudulent transactions represent a small minority. This study proposes a two-stage hybrid fraud detection model that integrates an Autoencoder for unsupervised anomaly detection with XGBoost as a supervised classifier. The objective is to evaluate whether incorporating reconstruction error scores from the Autoencoder as an additional feature improves XGBoost classification performance on highly imbalanced e-commerce fraud data. The dataset used is the Credit Card Fraud Detection dataset from Kaggle (ULB), consisting of 284,807 transactions with a fraud ratio of 0.17%. The research pipeline includes stratified train-validation-test splitting, StandardScaler normalization, Autoencoder training exclusively on non-fraud data to produce anomaly scores (AE_Score), and XGBoost training with scale_pos_weight to handle class imbalance. Threshold optimization was performed using the precision-recall curve on the validation set. Results demonstrate that the two-stage model achieved a ROC-AUC of 0.9749, PR-AUC of 0.8442, precision of 0.8971, recall of 0.8243, and F1-score of 0.8592 on the test set. The AE_Score showed strong discriminative power, with a fraud mean of 4.79 compared to 0.02 for non-fraud transactions. These findings confirm that integrating Autoencoder-based anomaly scoring into gradient boosting classification effectively addresses data imbalance and improves fraud detection performance in large-scale e-commerce environments.

Page 136 of 137 | Total Record : 1361

 133   134   135   136   137 

Filter by Year

2016 2026


Reset
Filter By Issues
All Issue Vol. 10 No. 3 (2026): Article Research July 2026 Vol. 10 No. 2 (2026): Article Research April, 2026 Vol. 10 No. 1 (2026): Article Research January 2026 Vol. 9 No. 4 (2025): Articles Research October 2025 Vol. 9 No. 3 (2025): Article Research July 2025 Vol. 9 No. 2 (2025): Research Articles April 2025 Vol. 9 No. 1 (2025): Research Article, January 2025 Vol. 8 No. 4 (2024): Article Research Volume 8 Issue 4, October 2024 Vol. 8 No. 3 (2024): Research Artikel Volume 8 Issue 3, July 2024 Vol. 8 No. 2 (2024): Article Research Volume 8 Issue 2, April 2024 Vol. 8 No. 1 (2024): Articles Research Volume 8 Issue 1, January 2024 Vol. 7 No. 4 (2023): Article Research Volume 7 Issue 4, October 2023 Vol. 7 No. 3 (2023): Article Research Volume 7 Issue 3, July 2023 Vol. 7 No. 2 (2023): Research Article, Volume 7 Issue 2 April, 2023 Vol. 7 No. 1 (2023): Articles Research Volume 7 Issue 1, 2023 Vol. 6 No. 4 (2022): Article Research: Volume 6 Number 4, October 2022 Vol. 6 No. 3 (2022): Article Research Volume 6 Number 3, July 2022 Vol. 6 No. 2 (2022): Articles Research Volume 6 Issue 2, April 2022 Vol. 6 No. 1 (2021): Article Research Volume 6 Issue 1: January 2021 Vol. 5 No. 2 (2021): Article Research Volume 5 Number 2, April 2021 Vol. 5 No. 2B (2021): Article Research October 2021 Vol 4 No 2 (2020): SinkrOn Volume 4 Number 2, April 2020 Vol. 5 No. 1 (2020): Article Research, October 2020 Vol. 4 No. 1 (2019): SinkrOn Volume 4 Number 1, October 2019 Vol 3 No 2 (2019): SinkrOn Volume 3 Number 2, April 2019 Vol. 3 No. 2 (2019): SinkrOn Volume 3 Number 2, April 2019 Vol. 3 No. 1 (2018): SinkrOn Volume 3 Nomor 1, Periode Oktober 2018 Vol 3 No 1 (2018): SinkrOn Volume 3 Nomor 1, Periode Oktober 2018 Vol. 2 No. 2 (2018): SinkrOn Volume 2 Nomor 2 April 2018 Vol. 2 No. 1 (2017): SinkrOn Volume 2 Nomor 1 Oktober 2017 Vol. 1 No. 2 (2017): SinkrOn Volume 1 Nomor 2 April 2017 Vol. 1 No. 1 (2016): SinkrOn Oktober Volume 1 Edisi 1 Tahun 2016 More Issue